Extended Methods
Total Page：16
File Type：pdf, Size：1020Kb
Extended Methods RNA extraction from tissue Skin tissue was homogenized in lysis buffer with Minilys by Bertin Technologies with 30sec homogenizing/30sec break cycle for a total of 3 cycles until tissue was completely disrupted. Alternatively, tissue was homogenized in Trizol (Thermo Fisher Scientific) using a tissue disruptor probe for 3 cycles of 30s pulses. Homogenates were processed for RNA extraction with Qiagen RNeasy Fibrous Tissue kit according to the manufacturer’s protocol. RNA concentration was assessed using a spectrophotometer (NanoDrop). RNA sequencing Skin biopsies of three different cohorts were used for RNA sequencing analysis (Cohort 1, PRESS and cohort 3). Total RNA was fractionated to purify mRNA using the Dynabeads mRNA Purification Kit (Thermo Fisher Scientific) to minimize contamination from rRNA. Purified polyA mRNA was used as the template for reverse transcription. After first strand complementary DNA (cDNA) synthesis, the entire single strand cDNA was used for the double strand cDNA synthesis. The MinElute PCR Purification Kit (Qiagen) was used to concentrate the cDNA, and size selection of 200-300bp cDNA was carried out on an agarose gel. Illumina sequencing libraries were generated for the cDNA according to the manufacturer’s sample preparation protocol, and then sequenced on the HiSeq2000 platform to generate paired 100bp long reads. Sequence reads was mapped to human genome using Bowtie and then processed with the software package TopHat and Cufflinks. The number of read counts per gene was estimated using HTseq. Differential expression analysis was carried out on variance stabilized counts using DEseq2 package 44 (Bioconductor, https://www.bioconductor.org/). In addition, clustering and principal component analyses were performed to screen for technical artefacts related to the date of sample collection, extraction and sequencing. These analyses did not indicate any batch effect. For the PRESS cohort similar procedures were pursued with the difference that a ribosomal reduction instead of poly (A) enrichment method was carried out. Specifically, cDNA libraries were prepared using the TruSeq Stranded Total RNA with Ribo-Zero Gold Kit (Illumina). Moreover, 2X 76 bp paired-end sequencing on HiSeq 2500 (Illumina) was conducted, generating approximately 50 million reads per sample. RNA sequencing performed on the skin biopsies from cohort 3 has been described recently (1). Protein coding capacity The LncRNA protein coding capacity was evaluated with the CPAT (http://rna-cpat.sourceforge.net/) analysis which is based on four pure sequence-based, linguistic features, ORF size, ORF coverage, Fickett TESTCODE and Hexamer usage bias. Analysis was done using default settings for human sequences, and coding probabilities below 0.364 were considered noncoding as described (2,3). Cells and cell culture Human pulmonary artery smooth muscle cells (HPASMC) were purchased from ScienCell and cultured in smooth muscle cell medium (SMCM, ScienCell) supplemented with 1% FBS, 100 units/ml of penicillin, 100 μg/ml of streptomycin and smooth muscle cell growth supplement (all ScienCell). Rheumatoid arthritis synovial fibroblasts (RASFs) were isolated from synovial tissues digested with dispase (37 °C, 1 h) and cultured in DMEM (Sigma-Aldrich) supplemented Penicillin and 50 μg/ml Streptomycin (Gibco) and 10% fetal bovine serum (10% FBS, Gibco), 2 mM L-glutamine, 10 mM HEPES and 0.2% amphotericin B (all from Life Technologies). BJ5TA hTERT-immortalized foreskin fibroblast cell line was purchased from ATCC and kept in medium constituted of four parts of DMEM (Sigma-Aldrich) and one part of Medium 199 (Sigma-Aldrich) supplemented with 0.01mg/ml hygromycin B (Sigma-Aldrich), 10% fetal bovine serum (10% FBS, Gibco), 50 U/ml Penicillin and 50μg/ml Streptomycin (Gibco). Human dermal microvascular endothelial cells (HDMEC) were purchased from ScienCell and cultured in endothelial cell medium (SMCM, ScienCell). CD14+ cells were isolated from blood of SSc patients using CD14 Microbeads (MACS, Miltenyi Biotec) according to the manufacturer’s protocol and cultured in DMEM (Sigma-Aldrich) containing 50 U/ml Penicillin and 50 μg/ml Streptomycin (Gibco) and 10% fetal bovine serum (10% FBS, Gibco) and 100 µM 2-mercaptoethanol (Gibco). Purity was confirmed to be >95% by FACS analysis. Primary human keratinocytes (HK) were isolated from healthy skin derived from donors undergoing plastic surgery as previously described (4). Keratinocytes were maintained in keratinocyte-serum-free medium (KSF-M, Gibco) supplied with epidermal growth factor (EGF, Gibco) and Bovine Pituitary Extract (BPE, Gibco). All cell types were maintained in 5% CO2 humid 37°C incubator. Cytokines and Inhibitors For stimulation experiments, starvation medium (1% FBS) was supplemented either with 0.1-10ng/ml TGFβ (Peprotech), 20ng/ml PDGF (Peprotech), 10ng/ml IL-1β (ImmunoTools), 10ng/ml IL-4 (ImmunoTools), 10ng/ml IL-13 (ImmunoTools) or 10ng/ml IL-17a (ImmunoTools) for 6, 24 or 72h. TGFβR1 was blocked using 100nM SD208 inhibitor (Tocris) and 10µM SB431542 inhibitor (Tocris) in parallel to TGFβ treatment. Cells were starved for 24h prior to stimulation. RNA extraction from cells Total RNA from cells was isolated with Zymo Quick-RNA MicroPrep RNA isolation kit. RNA concentration was assessed using a spectrophotometer (NanoDrop). Microarray Two-hundred ng of total RNA was amplified and purified using a TotalPrep RNA Amplification Kit (Applied Biosystems/Ambion). Reverse transcription for first strand cDNA was performed using a T7 Oligo(dT) Primer to synthesize cDNA containing a T7 promoter sequence. Single stranded cDNA was then converted into double stranded cDNA and purified. In vitro transcription was used to amplify and label multiple copies of biotinylated cRNA. The amplified cRNA was hybridized on Illumina HT-12 arrays. The R-package ‘lumi’ was used to read and process the microarray raw data. For differential expression analysis was performed using ‘limma’ (5). The annotation is based on the R-package ‘illuminaHumanv4.db’. Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) One hundred twenty-five ng of RNA and random hexamers were used to carry out reverse transcription using the Transcriptor First Strand cDNA Synthesis kit (Roche). Subsequent qPCRs were performed with 2x SYBR Green master mix (Promega) on an Agilent Technologies Stratagene Mx3005P qPCR system. Ribosomal protein lateral stalk subunit P0 (RPLP0) and Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) were used as reference housekeeping genes. Primer list is provided in Supplementary Table 7. To analyze the expression of microRNAs, 10ng of RNA were reverse-transcribed using TaqMan MicroRNA Reverse Transcription Kit (Thermo Fisher Scientific) and pre-designed loop specific primers for miR-424 and miR-503 (Assay ID: 000604, 001048, Thermo Fisher Scientific). Pre-designed single TaqMan miRNA assays (Assay ID: 000604, 001048, Thermo Fisher Scientific) and TaqMan Universal Master Mix II, no UNG (Thermo Fisher Scientific) were used to measure the expression levels of miR-424 and miR-503. Expression of U48 small nucleolar RNA (RNU48) was used as endogenous control (Assay ID: 001006, Thermo Fisher Scientific). Murine models For bleomycin induced fibrosis, female 8 week old C57Bl6/J mice (Janvier, Le Genest-Saint-Isle, France) were injected intradermally every other day with either 100 bleomycin (0.5-1 mg/ml) or the vehicle NaCl (0.9%) (6) for 4 weeks. Lung fibrosis was induced in female 8 week old C57Bl6/J mice (Janvier, Le Genest-Saint-Isle, France) using intratracheally instilled bleomycin sulfate (Baxter, Kantonsapotheke Zurich, Switzerland) at a dosage of 4-2 U/kg of body weight (7). Control mice received equivalent volumes (50 μl) of 0.9% NaCl solution. For AdTGFβR1-induced fibrosis, 12-week-old mice (C57BL/6 background, Janvier, Le Genest-Saint-Isle, France, mixed genders) received of 6.67 × 107 pfu/mouse of replication-deficient type 5 adenoviruses encoding for constitutively active TGFβR1 into marked areas of the upper back every other week for two months. Injections of replication-deficient type 5 adenoviruses encoding for LacZ were used as controls (8). TSK‐1 mice (10 weeks of age, mixed genders) were interbred with pa/pa mice ( The Jackson Laboratory, Bar Harbor, ME, USA).TSK-1 mice spontaneously develop increased dermal and hypodermal thickness as result of a partial in- frame duplication in the fibrillin-1 gene. Bioinformatics analysis for ATAC-seq and ChIRP data Data analysis for ChIRP-seq and ATAC-seq experiments was performed using the data analysis framework SUSHI (9). The raw reads were first cleaned by removing adapter sequences, trimming low quality ends, and filtering reads with low quality (phred quality <20) using Trimmomatic (10). Sequence alignment of the resulting high-quality reads to the Human genome (build GRCh38) was performed using Bowtie2 (Version 2.3.2) (11) with non-default options for the ATAC-Seq data: --no-mixed --no-discordant --very-sensitive -X 1500. For peak calling we used MACS2 (12). The resulting peaks were merged using Bedtools. The peak annotation to the closest transcription start site (TSS) was performed with the R-package ChIPpeakAnno using Ensembl’s gene models of release 91 downloaded on February 26, 2018. The count values were computed with the function featureCounts from the R package R subread. Differential Count analysis was performed with the software package EdgeR (13). Database search WashU Epigenome Browser (http://epigenomegateway.wustl.edu/browser/)